A NOVEL BLIND DECONVOLUTION SCHEME FOR IMAGE-RESTORATION USING RECURSIVE FILTERING

In this paper, we present a novel blind deconvolution technique for th e restoration of linearly degraded images without explicit knowledge o f either the original image or the point spread function, The techniqu e applies to situations in which the scene consists of a finite suppor t object against a uniformly black, grey, or white background, This oc curs in certain types of astronomical imaging, medical imaging, and on e-dimensional (1-D) gamma ray spectra processing, among others, The on ly information required are the nonnegativity of the true image and th e support size of the original object. The restoration procedure invol ves recursive filtering of the blurred image to minimize a convex cost function, We prove convexity of the cost function, establish sufficie nt conditions to guarantee a unique solution, and examine the performa nce of the technique in the presence of noise, The new approach is exp erimentally shown to be more reliable and to have faster convergence t han existing nonparametric finite support blind deconvolution methods, For situations in which the exact object support is unknown, we propo se a novel support-finding algorithm.